metadata
license: mit
tags:
- generated_from_trainer
datasets:
- ncbi_disease
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: biogpt-new
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: ncbi_disease
type: ncbi_disease
config: ncbi_disease
split: validation
args: ncbi_disease
metrics:
- name: Precision
type: precision
value: 0.4968085106382979
- name: Recall
type: recall
value: 0.5933926302414231
- name: F1
type: f1
value: 0.540822235089751
- name: Accuracy
type: accuracy
value: 0.9570189427826831
biogpt-new
This model is a fine-tuned version of microsoft/biogpt on the ncbi_disease dataset. It achieves the following results on the evaluation set:
- Loss: 0.1984
- Precision: 0.4968
- Recall: 0.5934
- F1: 0.5408
- Accuracy: 0.9570
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1683 | 1.0 | 680 | 0.1669 | 0.3778 | 0.4752 | 0.4209 | 0.9512 |
| 0.1407 | 2.0 | 1360 | 0.1529 | 0.4183 | 0.5337 | 0.4690 | 0.9521 |
| 0.0813 | 3.0 | 2040 | 0.1548 | 0.4751 | 0.5820 | 0.5231 | 0.9570 |
| 0.0592 | 4.0 | 2720 | 0.1762 | 0.4966 | 0.5565 | 0.5249 | 0.9582 |
| 0.051 | 5.0 | 3400 | 0.1984 | 0.4968 | 0.5934 | 0.5408 | 0.9570 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3